netdev - Re: Vmxnet3 v22

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Fri, 20 Oct 2023 23:13:55 +0300
From: Martin Zaharinov <micron10@...il.com>
To: William Tu <u9012063@...il.com>
Cc: Alexander Duyck <alexanderduyck@...com>,
 alexandr.lobakin@...el.com,
 netdev <netdev@...r.kernel.org>,
 Sankararaman Jayaraman <jsankararama@...are.com>,
 doshir@...are.com,
 Boon Ang <bang@...are.com>
Subject: Re: Vmxnet3 v22 - bug

Hi William,

Big sorry for delay,
all work for now and don’t see any error.
its ok to commit.


If i see any other error i will update you .

Thanks for you great work.

Best regrads,
Martin

> On 14 Oct 2023, at 16:53, William Tu <u9012063@...il.com> wrote:
> 
> Hi Martin,
> Does everything work ok, do you find more issue?
> Otherwise I can send out patch to netdev, thanks!
> William
> 
> On Wed, Oct 4, 2023 at 1:33 AM Martin Zaharinov <micron10@...il.com> wrote:
>> 
>> Hi William,
>> 
>> Yes this patch fix problem ,
>> 
>> i will make little test if see any problem will update you.
>> 
>> Thanks for your great work!
>> 
>> Martin
>> 
>>> On 30 Sep 2023, at 20:03, William Tu <u9012063@...il.com> wrote:
>>> 
>>> On Fri, Sep 29, 2023 at 1:30 PM William Tu <u9012063@...il.com> wrote:
>>>> 
>>>> On Mon, Sep 4, 2023 at 9:24 AM Martin Zaharinov <micron10@...il.com> wrote:
>>>>> 
>>>>> Hi William Tu
>>>>> 
>>>>> 
>>>>> this is report of bug with latest version of vmxnet3 xdp support:
>>>>> 
>>>>> 
>>>>> [   92.417855] ------------[ cut here ]------------
>>>>> [   92.417855] XDP_WARN: xdp_update_frame_from_buff(line:278): Driver BUG: missing reserved tailroom
>>>>> [   92.417855] WARNING: CPU: 0 PID: 0 at net/core/xdp.c:586 xdp_warn+0xf/0x20
>>>>> [   92.417855] Modules linked in:  pppoe pppox ppp_generic slhc virtio_net net_failover failover virtio_pci virtio_pci_legacy_dev virtio_pci_modern_dev virtio virtio_ring vmxnet3
>>>>> [   92.417855] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G        W  O       6.5.1 #1
>>>>> [   92.417855] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.16.2-0-gea1b7a073390-prebuilt.qemu.org 04/01/2014
>>>>> [   92.417855] RIP: 0010:xdp_warn+0xf/0x20
>>>>> [   92.417855] Code: 00 00 c3 0f 1f 84 00 00 00 00 00 83 7f 0c 01 0f 94 c0 c3 0f 1f 84 00 00 00 00 00 48 89 f9 48 c7 c7 3d b2 e4 91 e8 d1 00 8e ff <0f> 0b c3 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 53 48 89 fb 8b
>>>>> [   92.417855] RSP: 0018:ffffb30180003d40 EFLAGS: 00010286
>>>>> [   92.417855] RAX: 0000000000000055 RBX: ffff99bcf7c22ee0 RCX: 00000000fffdffff
>>>>> [   92.417855] RDX: 00000000fffdffff RSI: 0000000000000001 RDI: 00000000ffffffea
>>>>> [   92.417855] RBP: ffff99bb849c2000 R08: 0000000000000000 R09: 00000000fffdffff
>>>>> [   92.417855] R10: ffff99bcf6a00000 R11: 0000000000000003 R12: ffff99bb83840000
>>>>> [   92.417855] R13: ffff99bb83842780 R14: ffffb3018081d000 R15: ffff99bb849c2000
>>>>> [   92.417855] FS:  0000000000000000(0000) GS:ffff99bcf7c00000(0000) knlGS:0000000000000000
>>>>> [   92.417855] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>>>>> [   92.417855] CR2: 00007f9bf822df88 CR3: 00000001a74de000 CR4: 00000000003506f0
>>>>> [   92.417855] Call Trace:
>>>>> [   92.417855]  <IRQ>
>>>>> [   92.417855]  ? __warn+0x6c/0x130
>>>>> [   92.417855]  ? report_bug+0x1e4/0x260
>>>>> [   92.417855]  ? handle_bug+0x36/0x70
>>>>> [   92.417855]  ? exc_invalid_op+0x17/0x1a0
>>>>> [   92.417855]  ? asm_exc_invalid_op+0x16/0x20
>>>>> [   92.417855]  ? xdp_warn+0xf/0x20
>>>>> [   92.417855]  xdp_do_redirect+0x15f/0x1c0
>>>>> [   92.417855]  vmxnet3_run_xdp+0x17a/0x400 [vmxnet3]
>>>>> [   92.417855]  vmxnet3_process_xdp+0xe4/0x760 [vmxnet3]
>>>>> [   92.417855]  ? vmxnet3_tq_tx_complete.isra.0+0x21e/0x2c0 [vmxnet3]
>>>>> [   92.417855]  vmxnet3_rq_rx_complete+0x7ad/0x1120 [vmxnet3]
>>>>> [   92.417855]  vmxnet3_poll_rx_only+0x2d/0xa0 [vmxnet3]
>>>>> [   92.417855]  __napi_poll+0x20/0x180
>>>>> [   92.417855]  net_rx_action+0x177/0x390
>>>>> [   92.417855]  __do_softirq+0xd0/0x202
>>>>> [   92.417855]  irq_exit_rcu+0x82/0xa0
>>>>> [   92.417855]  common_interrupt+0x7a/0xa0
>>>>> [   92.417855]  </IRQ>
>>>>> [   92.417855]  <TASK>
>>>>> [   92.417855]  asm_common_interrupt+0x22/0x40
>>>>> [   92.417855] RIP: 0010:default_idle+0xb/0x10
>>>>> [   92.417855] Code: 07 76 e7 48 89 07 49 c7 c0 08 00 00 00 4d 29 c8 4c 01 c7 4c 29 c2 e9 72 ff ff ff cc cc cc cc eb 07 0f 00 2d 47 72 29 00 fb f4 <fa> c3 0f 1f 00 65 48 8b 04 25 00 33 02 00 f0 80 48 02 20 48 8b 10
>>>>> [   92.417855] RSP: 0018:ffffffff92003e88 EFLAGS: 00000206
>>>>> [   92.417855] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000001
>>>>> [   92.417855] RDX: 4000000000000000 RSI: 0000000000000083 RDI: 00000000000bfc34
>>>>> [   92.417855] RBP: ffffffff92009dc0 R08: ffff99bcf7c1f160 R09: ffff99bcf7c1f100
>>>>> [   92.417855] R10: ffff99bcf7c1f100 R11: 0000000000000000 R12: 0000000000000000
>>>>> [   92.417855] R13: 0000000000000000 R14: ffffffff92009dc0 R15: 0000000000000000
>>>>> [   92.417855]  default_idle_call+0x1f/0x30
>>>>> [   92.417855]  do_idle+0x1df/0x210
>>>>> [   92.417855]  cpu_startup_entry+0x14/0x20
>>>>> [   92.417855]  rest_init+0xc7/0xd0
>>>>> [   92.417855]  arch_call_rest_init+0x5/0x20
>>>>> [   92.417855]  start_kernel+0x3e9/0x5b0
>>>>> [   92.417855]  x86_64_start_reservations+0x14/0x30
>>>>> [   92.417855]  x86_64_start_kernel+0x71/0x80
>>>>> [   92.417855]  secondary_startup_64_no_verify+0x167/0x16b
>>>>> [   92.417855]  </TASK>
>>>>> [   92.417855] ---[ end trace 0000000000000000 ]—
>>>>> 
>>>>> 
>>>> Hi Martin,
>>>> 
>>>> Thanks, I'll take a look.
>>>> William
>>> 
>>> Hi Martin,
>>> For non-dataring packet, I should use rbi->len instead of rcd->len.
>>> Are you able to see if this fixes the bug?
>>> thanks!
>>> 
>>> diff --git a/drivers/net/vmxnet3/vmxnet3_xdp.c
>>> b/drivers/net/vmxnet3/vmxnet3_xdp.c
>>> index 80ddaff759d4..a6c787454a1a 100644
>>> --- a/drivers/net/vmxnet3/vmxnet3_xdp.c
>>> +++ b/drivers/net/vmxnet3/vmxnet3_xdp.c
>>> @@ -382,12 +382,12 @@ vmxnet3_process_xdp(struct vmxnet3_adapter *adapter,
>>>       page = rbi->page;
>>>       dma_sync_single_for_cpu(&adapter->pdev->dev,
>>>                               page_pool_get_dma_addr(page) +
>>> -                               rq->page_pool->p.offset, rcd->len,
>>> +                               rq->page_pool->p.offset, rbi->len,
>>>                               page_pool_get_dma_dir(rq->page_pool));
>>> 
>>> -       xdp_init_buff(&xdp, rbi->len, &rq->xdp_rxq);
>>> +       xdp_init_buff(&xdp, PAGE_SIZE, &rq->xdp_rxq);
>>>       xdp_prepare_buff(&xdp, page_address(page), rq->page_pool->p.offset,
>>> -                        rcd->len, false);
>>> +                        rbi->len, false);
>>>       xdp_buff_clear_frags_flag(&xdp);
>>> 
>>>       xdp_prog = rcu_dereference(rq->adapter->xdp_bpf_prog);
>>